By Louis Simoneau

Non-Latin TLDs Spotted in the Wild

By Louis Simoneau

At the end of last year, ICANN decided to begin accepting applications for country code top-level domains (TLDs) in non-Latin scripts. At the time, they admitted it was only a first step, and that much work remained to be done. Last week, however, the first live examples of non-Latin TLDs were turned on. The domains in question are for Egypt: مصر (Egypt), Saudi Arabia: السعودية (AlSaudiah), and the United Arab Emirates: امارات (Emarat). Those domains support URLs being written entirely from right to left (so that the domain would come first).

As an example, a website that’s already using one of the new domains is http://وزارة-الأتصالات.مصر/ (it’s hard to see, but you should be able to make out the period that comes after the first few characters, as the URL is written right to left). This URL should lead you to the Egyptian Ministry of Communications and Information Technology’s website.

The newest versions of all the major browsers are capable of handling internationalized domain names (IDNs). Your browser should load the correct site, though it might not display the Arabic URL in the address bar; it does depend on the domain whitelist and language settings of your browser. ICANN has helpfully posted a fairly detailed explanation of how to make the new domains look pretty in IE8, Firefox, and Opera.

In addition to the three Arabic domains mentioned above, ICANN has also approved the рф (rf) domain for the Russian Federation, with further work to occur at the Russian Internet Governance Forum that takes place this week.

I think we can all agree that this is a great development for the Web, helping to make it more accessible and usable to a much wider range of languages and cultures. Congrats to ICANN and these countries on a job well done!

  • Ok… nice feature and all… but what’s the point of having tlds that few would be able to read?

    Wouldn’t it be a phishing issue as well?

    • So only few can read the domains and only those few can read the site anyways. So…yeah what is the problem? Or do you make it a habit going to sites written in a language you don’t know?

  • Gryffyn

    I still want to know how this is going to affect current punycode oriented domains. Punycode is a method of transforming unicode characters into latin characters in a predictable manner, so you can register latin character garbage looking URLs but that most browsers will translate a proper looking unicode URL into and access correctly (you get redirected to the ugly URL but still).

    Look at as an example of an extreme URL shortener that uses punycode to give you, sometimes literally, 2-3 character URLs (minus the .ws).

    When does a browser decide to convert punycode to latin versus access the unicode URL directly?

  • Dusan

    I am from Serbia and our official letter is Cyrillic. That stated, I think that this is just unnecessary complication and that will not help anyone.

    Imho, internationalization should happen in “national pride” heads.
    English set itself, with no artificial force, as international language no 1, so learn it and internationalize yourself.

    Anyway, can’t wait to see how will this change the internet and it’s usage we know.


  • Jacques

    In bowing to outside forces, ICANN has opened an incredibly dangerous can of worms. Security concerns thought tracking / tracing urls based on a 26-character alphabet (essentially) was difficult? Now they will have to deal with alphabets that have literally tens of thousands of distinct characters in one well-known case. This will make tracking terrorists and other global miscreants incredibly more difficult and will most certainly lead to more casualties on all sides. I have friends on every continent but we save our native languages for the content not the header. American English is not hard to learn. They prove that every day, eh! ;-)

    • I don’t know where to begin…but that is just silly what you are saying. This is not going to effect the tracking of “terrorists” in any shape or form. There are no security implication in this, NONE.

  • Weirdan

    > I think we can all agree that this is a great development for the Web
    Really? What’s so good about it?

    It was bad enough when IDN was introduced – I had to clarify on several occasions that users needed to use Cyrillic in part of the domain. Now there will be latin-trasliterated domains, old-style IDNs (where tld is in latin) AND new-style IDNs (with tld in local scripts).

  • arts-multimedia

    The internet is about bringing people together and in order to do that, you need a universal language everybody more or less understands.
    English is for computers what Latin is for doctors and I find therefore the idea of diversifying domain names into different languages a stupid idea.

    I also hate it when google sends me automatically to just because I happen to live in Belgium, or even worse, facebook addressing me in French while I don’t even understand that language.
    Please, let us keep a universal language! It’s the only way to talk with everyone worldwide!

    • The universal language is not English. It is math.

  • Enisa

    There are localized all sorts of software, you can google in your own language, etc. so why not this?
    Don’t get me wrong, for many sites, maybe for most of them, this is useless. But, it is nice for sites that are for local audience in countries where the Roman alphabet is not used. More people will be able to use the internet in more ways.
    @Dusan: it’s easy for us, because Serbian is also written in Latin alphabet. Cyrillic is official, but both are recognised and we learn both in school. There’s a standard way to transliterate between the two. It’s very different for people who do not use Latin script in their language.

  • This is a great feature. More people who doesn’t know English will be able to use the internet!

  • andarin2

    IMHO this is a very unfortunate move….
    on the one hand, it is not that useful to have new top level domains… which will only be usable -mostly- within that country itself.

    even worse, examples show that these new domains may conflict (at least conceptually) with existing ones. for instance Russia is .рф, which is transliterated to .rf, different from .ru.

    moreover this is a step towards mutual isolation, encouraging internal communication within each culture…

Get the latest in Front-end, once a week, for free.